Event information may be included in unstructured natural language information. For example, two users may use text messages to arrange a lunch event. Once the users agree on the event, they may wish to create a calendar entry for the event. Conventionally, to create a calendar entry, a user is required to manually transfer event information from unstructured natural language such as a text message to a calendar entry. Manually transferring event information typically requires multiple copy and paste operations or tedious manual entry. The process of manually creating the calendar entry is thus cumbersome and time consuming. Accordingly, there is a need for more efficient methods for detecting and classifying events based on unstructured natural language information.